Table 7 Technological needs and challenges for space robotics in the coming decades.
AreasGoalsTechnological needs or challengesRelevance to achieving top-level science
Sensing and
To provide situational
awareness for space
robotic agents,
explorers, and
- New sensors
- Sensing techniques
- Algorithms for 3D perception, state estimation,
and data fusion
- Onboard data processing and generic software
- Object, event, or activity recognition
The sensors provide the vast bulk of the direct
-Increases in instruments, both remote sensing and
in situ enable more precise measurements (e.g.,
spatial, spectral resolution, while reducing
volume, mass, and power).
- New types of instruments are emerging. Imaging
spectroscopy to determine composition; lidar for
3D mapping; interferometric radar for change
detection, structure; sample processing for life
detection and astrobiology to enable new
measurements for new types of science.
Mobility or locomotionTo reach and operate
at sites of scientific
interest on
surfaces or free
space environments
- Mobility on, into, and above an extraterrestrial
surface using locomotion like flying, walking,
climbing, rappelling, tunneling, swimming,
and sailing
- Melting through the kilometers-thick ocean
worlds’ ice shells of Europa, Enceladus, or Pluto
- Manipulations to make intentional changes in
the environment or objects using locomotion
like placing, assembling, digging, trenching,
drilling, sampling, grappling, and berthing
Locomotion represents the ability to explore an
environment, such as rovers, aerobots, and
submarines. Melting through ocean worlds’ ice
shells enables access to habitable oceans
underneath. Digging, trenching, and coring enable
access to materials without atmospheric
contamination (e.g., Mars geology) or radiation
(e.g., Europa astrobiology).
High-level autonomy
for system and
To provide robust and
safe autonomous
rendezvous, and
docking capabilities
and to enable
operations without
human interventions
to improve overall
performance of
human and robotic
missions. To enable
closed-loop science
for more efficient,
novel science (e.g.,
tracking a dynamic
plume at a comet)
- GNC algorithms
- Docking and capture mechanisms and interfaces
- Planning, scheduling, and common autonomy
software framework
- Multi-agent coordination
- Reconfigurable and adjustable autonomy
- Automated data analysis for decision-making,
fault detection, isolation and recovery/IVHM,
and execution
- Enhanced guidance navigation and control means
higher precision navigation for better science
measurements. Scheduling, execution, and
integrated vehicle health management enable
more productive science time for vehicles.
- Automated science analysis and scheduling enable
closing the loop without ground in the loop,
enabling more science cycles per mission (i.e.,
higher productivity and unique, opportunistic
To enable humans to
accurately and
rapidly understand
the state of the
robot in
collaboration and
act effectively and
efficiently toward
the goal state
- Multimodal interaction; remote and supervised
- Proximate interaction
- Distributed collaboration and coordination
- Common human-system interfaces
Virtual reality and augmented reality allow more
natural interfaces to analyze vast acquired data
streams. Virtual reality and augmented reality also
allow for natural means of vehicle controlling such
as by reach, touch, and gesture.
System engineeringTo provide a
framework for
understanding and
coordinating the
interactions of
robots and
achieving the
desired system
- Modularity, commonality, and interfaces
- Verification and validation of complex adaptive
- Robot modeling and simulation
- Software architectures and frameworks
- Safety and trust
High stakes in billions require a reliable mission. As
systems become increasingly complex, being able
to characterize robotic behavior (especially for
multivehicle swarms) becomes increasingly